Opioid-induced rewards, locomotion, and dopamine activation: A proposed model for control by mesopontine and rostromedial tegmental neurons.

Opioids, such as morphine or heroin, increase forebrain dopamine (DA) release and locomotion, and support the acquisition of conditioned place preference (CPP) or self-administration. The most sensitive sites for these opioid effects in rodents are in the ventral tegmental area (VTA) and rostromedial tegmental nucleus (RMTg). Opioid inhibition of GABA neurons in these sites is hypothesized to lead to arousing and rewarding effects through disinhibition of VTA DA neurons. We review findings that the laterodorsal tegmental (LDTg) and pedunculopontine tegmental (PPTg) nuclei, which each contain cholinergic, GABAergic, and glutamatergic cells, are important for these effects. LDTg and/or PPTg cholinergic inputs to VTA mediate opioid-induced locomotion and DA activation via VTA M5 muscarinic receptors. LDTg and/or PPTg cholinergic inputs to RMTg also modulate opioid-induced locomotion. Lesions or inhibition of LDTg or PPTg neurons reduce morphine-induced increases in forebrain DA release, acquisition of morphine CPP or self-administration. We propose a circuit model that links VTA and RMTg GABA with LDTg and PPTg neurons critical for DA-dependent opioid effects in drug-naïve rodents.

Muscarinic control of rostromedial tegmental nucleus GABA neurons and morphine-induced locomotion.

Opioids induce rewarding and locomotor effects by inhibiting rostromedial tegmental GABA neurons that express µ-opioid and nociceptin receptors. These GABA neurons then strongly inhibit dopamine neurons. Opioid-induced reward, locomotion and dopamine release also depend on pedunculopontine and laterodorsal tegmental cholinergic and glutamate neurons, many of which project to and activate ventral tegmental area dopamine neurons. Here we show that laterodorsal tegmental and pedunculopontine cholinergic neurons project to both rostromedial tegmental nucleus and ventral tegmental area, and that M4 muscarinic receptors are co-localized with µ-opioid receptors associated with rostromedial tegmental GABA neurons. To inhibit or excite rostromedial tegmental GABA neurons, we utilized adeno-associated viral vectors and DREADDs to express designed muscarinic receptors (M4D or M3D respectively) in GAD2::Cre mice. In M4D-expressing mice, clozapine-N-oxide increased morphine-induced, but not vehicle-induced, locomotion. In M3D-expressing mice, clozapine-N-oxide blocked morphine-induced, but not vehicle-induced, locomotion. We propose that cholinergic inhibition of rostromedial tegmental GABA neurons via M4 muscarinic receptors facilitates opioid inhibition of the same neurons. This model explains how mesopontine cholinergic systems and muscarinic receptors in the rostromedial tegmental nucleus and ventral tegmental area are important for dopamine-dependent and dopamine-independent opioid-induced rewards and locomotion.

Parvalbumin and GAD65 interneuron inhibition in the ventral hippocampus induces distinct behavioral deficits relevant to schizophrenia.

Hyperactivity within the ventral hippocampus (vHPC) has been linked to both psychosis in humans and behavioral deficits in animal models of schizophrenia. A local decrease in GABA-mediated inhibition, particularly involving parvalbumin (PV)-expressing GABA neurons, has been proposed as a key mechanism underlying this hyperactive state. However, direct evidence is lacking for a causal role of vHPC GABA neurons in behaviors associated with schizophrenia. Here, we probed the behavioral function of two different but overlapping populations of vHPC GABA neurons that express either PV or GAD65 by selectively inhibiting these neurons with the pharmacogenetic neuromodulator hM4D. We show that acute inhibition of vHPC GABA neurons in adult mice results in behavioral changes relevant to schizophrenia. Inhibiting either PV or GAD65 neurons produced distinct behavioral deficits. Inhibition of PV neurons, affecting ∼80% of the PV neuron population, robustly impaired prepulse inhibition of the acoustic startle reflex (PPI), startle reactivity, and spontaneous alternation, but did not affect locomotor activity. In contrast, inhibiting a heterogeneous population of GAD65 neurons, affecting ∼40% of PV neurons and 65% of cholecystokinin neurons, increased spontaneous and amphetamine-induced locomotor activity and reduced spontaneous alternation, but did not alter PPI. Inhibition of PV or GAD65 neurons also produced distinct changes in network oscillatory activity in the vHPC in vivo. Together, these findings establish a causal role for vHPC GABA neurons in controlling behaviors relevant to schizophrenia and suggest a functional dissociation between the GABAergic mechanisms involved in hippocampal modulation of sensorimotor processes.

Cholinergic control of morphine-induced locomotion in rostromedial tegmental nucleus versus ventral tegmental area sites.

M5 muscarinic acetylcholine receptors expressed on ventral tegmental dopamine (DA) neurons are needed for opioid activation of DA outputs. Here, the M5 receptor gene was bilaterally transfected into neurons in the ventral tegmental area (VTA) or the adjacent rostromedial tegmental nucleus (RMTg) in mice by means of a Herpes simplex viral vector (HSV) to increase the effect of endogenous acetylcholine. Three days after HSV-M5 gene infusion in VTA sites, morphine-induced locomotion more than doubled at two doses, while saline-induced locomotion was unaffected. When the HSV-M5 gene was infused into the adjacent RMTg, morphine-induced locomotion was strongly inhibited. The sharp boundary between these opposing effects was found where tyrosine hydroxylase (TH) and cholinesterase labelling decreases (-4.00 mm posterior to bregma). The same HSV-M5 gene transfections in M5 knockout mice induced even stronger inhibitory behavioural effects in RMTg but more variability in VTA sites due to stereotypy. The VTA sites where HSV-M5 increased morphine-induced locomotion receive direct inputs from many RMTg GAD-positive neurons, and from pontine ChAT-positive neurons, as shown by cholera-toxin B retrograde tracing. Therefore, morphine-induced locomotion was decreased by M5 receptor gene expression in RMTg GABA neurons that directly inhibit VTA DA neurons. Conversely, enhancing M5 receptor gene expression on VTA DA neurons increased morphine-induced locomotion via cholinergic inputs.

Acute food deprivation reverses morphine-induced locomotion deficits in M5 muscarinic receptor knockout mice.

Lesions of the pedunculopontine tegmental nucleus (PPT), one of two sources of cholinergic input to the ventral tegmental area (VTA), block conditioned place preference (CPP) for morphine in drug-naïve rats. M5 muscarinic cholinergic receptors, expressed by midbrain dopamine neurons, are critical for the ability of morphine to increase nucleus accumbens dopamine levels and locomotion, and for morphine CPP. This suggests that M5-mediated PPT cholinergic inputs to VTA dopamine neurons critically contribute to morphine-induced dopamine activation, reward and locomotion. In the current study we tested whether food deprivation, which reduces PPT contribution to morphine CPP in rats, could also reduce M5 contributions to morphine-induced locomotion in mice. Acute 18-h food deprivation reversed the phenotypic differences usually seen between non-deprived wild-type and M5 knockout mice. That is, food deprivation increased morphine-induced locomotion in M5 knockout mice but reduced morphine-induced locomotion in wild-type mice. Food deprivation increased saline-induced locomotion equally in wild-type and M5 knockout mice. Based on these findings, we suggest that food deprivation reduces the contribution of M5-mediated PPT cholinergic inputs to the VTA in morphine-induced locomotion and increases the contribution of a PPT-independent pathway. The contributions of cholinergic, dopaminergic and GABAergic neurons to the effects of acute food deprivation are discussed.

Muscarinic receptors in brain stem and mesopontine cholinergic arousal functions.

All five muscarinic receptor subtypes and mRNAs are found widely in the brain stem, with M2 muscarinic receptors most concentrated in the hindbrain. Three cholinergic cell groups, Ch5: pedunculopontine (PPT); Ch6: laterodorsal tegmental (LDT); Ch8: parabigeminal (PBG), are found in the tegmentum. Ch5,6 neurons are activated by arousing and reward-activating stimuli, and inhibited via M2-like autoreceptors. Ch5,6 ascending projections activate many forebrain regions, including thalamus, basal forebrain, and orexin/hypocretin neurons (via M3 receptors) for waking arousal and attention. Ch5,6 activation of dopamine neurons of the ventral tegmental area and substantia nigra (via M5 receptors) increases reward-seeking and energizes motor functions. M5 receptors on dopamine neurons facilitate brain-stimulation reward, opiate rewards and locomotion, and male ultrasonic vocalizations during mating in rodents. Ch5 cholinergic activation of superior colliculus intermediate layers facilitates fast saccades and approach turns, accompanied by nicotinic and muscarinic inhibition of the startle reflex in pons. Ch8 PBG neurons project to the outer layers of the superior colliculus only, where M2 receptors are associated with retinotectal terminals. Ch5,6 descending projections to dorsal pontine reticular formation contribute to M2-dependent REM sleep.

M5 muscarinic receptors mediate striatal dopamine activation by ventral tegmental morphine and pedunculopontine stimulation in mice.

Opiates, like other addictive drugs, elevate forebrain dopamine levels and are thought to do so mainly by inhibiting GABA neurons near the ventral tegmental area (VTA), in turn leading to a disinhibition of dopamine neurons. However, cholinergic inputs from the laterodorsal (LDT) and pedunculopontine (PPT) tegmental nucleus to the VTA and substantia nigra (SN) importantly contribute, as either LDT or PPT lesions strongly attenuate morphine-induced forebrain dopamine elevations. Pharmacological blockade of muscarinic acetylcholine receptors in the VTA or SN has similar effects. M5 muscarinic receptors are the only muscarinic receptor subtype associated with VTA and SN dopamine neurons. Here we tested the contribution of M5 muscarinic receptors to morphine-induced dopamine elevations by measuring nucleus accumbens dopamine efflux in response to intra-VTA morphine infusion using in vivo chronoamperometry. Intra-VTA morphine increased nucleus accumbens dopamine efflux in urethane-anesthetized wildtype mice starting at 10 min after infusion. These increases were absent in M5 knockout mice and were similarly blocked by pre-treatment with VTA scopolamine in wildtype mice. Furthermore, in wildtype mice electrical stimulation of the PPT evoked an initial, short-lasting increase in striatal dopamine efflux, followed 5 min later by a second prolonged increase in dopamine efflux. In M5 knockout mice, or following systemic pre-treatment with scopolamine in wildtype mice, the prolonged increase in striatal dopamine efflux was absent. The time course of increased accumbal dopamine efflux in wildtype mice following VTA morphine was consistent with both the prolonged M5-mediated excitation of striatal dopamine efflux following PPT electrical stimulation and accumbal dopamine efflux following LDT electrical stimulation. Therefore, M5 receptors appear critical for prolonged PPT excitation of dopamine efflux and for dopamine efflux induced by intra-VTA morphine.

GABA receptors and prepulse inhibition of acoustic startle in mice and rats.

The acoustic startle reflex is strongly inhibited by a moderate-intensity acoustic stimulus that precedes the startling stimulus by roughly 10-1000 ms (prepulse inhibition, PPI). At long interstimulus intervals (ISIs) of 100-1000 ms, PPI in rats is reduced by the muscarinic receptor antagonist scopolamine. Here, we studied the role of GABA receptors in PPI at full ISI ranges in both mice and rats. In B6 mice, PPI begins and ends at shorter ISIs (4 and 1000 ms, respectively) than in Wistar rats (8 and 5000 ms). The GABA(A) antagonist bicuculline (1 mg/kg i.p.) reduced PPI at ISIs near the peak of PPI in both rats and mice. The GABA(B) antagonist phaclofen (10 or 30 mg/kg i.p. in rats or mice, respectively) reduced PPI only at long ISIs, similar to the effects of the muscarinic antagonist scopolamine (1 mg/kg i.p.). The effects of phaclofen and scopolamine were additive in rats, suggesting independent effects of GABA(B) and muscarinic receptors. Patch-clamp recordings of startle-mediating PnC (nucleus reticularis pontis caudalis) giant neurons in rat slices show that EPSCs evoked by either trigeminal or auditory fiber stimulation were inhibited by the GABA(A/C) agonist muscimol or the GABA(B) agonist baclofen via postsynaptic mechanisms. Hyperpolarization of PnC neurons by muscimol was reversed with bicuculline, indicating that postsynaptic GABA(A) receptors strongly inhibit PnC giant neurons needed for startle. Therefore, GABA receptors on PnC giant neurons mediate a substantial part of PPI, with GABA(A) receptors contributing at the peak of PPI, and GABA(B) receptors adding to muscarinic effects on PPI at long ISIs.

M5 muscarinic receptor knockout mice show reduced morphine-induced locomotion but increased locomotion after cholinergic antagonism in the ventral tegmental area.

M(5) muscarinic receptors are the only muscarinic receptor subtype expressed by mesencephalic dopamine neurons and provide an important excitatory input to mesolimbic and nigrostriatal dopamine systems. Here, we studied locomotion induced by systemic morphine (3, 10, and 30 mg/kg i.p.) in M(5) knockout mice of the C57BL/6 (B6) and CD1 x 129SvJ background strains. M(5) knockout mice of both strains showed reduced locomotion in response to 30 mg/kg morphine. B6 M(5) knockout mice were less sensitive to naltrexone in either the antagonism of morphine-induced locomotion or in the reduction of locomotion by naltrexone alone. This suggests that M(5) knockout mice are less sensitive to the effects of either exogenous or endogenous opiates on locomotion and that spontaneous locomotion in B6 mice is sustained by endogenous opiates. In B6 wild-type mice, ventral tegmental area (VTA) pretreatment with the muscarinic receptor antagonist atropine (3 microg bilateral), but not the nicotinic receptor antagonist mecamylamine (5 microg bilateral), reduced locomotion in response to 30 mg/kg morphine to a similar extent as systemic M(5) knockout, suggesting that reduced morphine-induced locomotion in M(5) knockout mice is due to the loss of M(5) receptors on VTA dopamine neurons. In contrast, in M(5) knockout mice, but not in wild-type mice, either intra-VTA atropine or mecamylamine alone increased locomotion by almost 3 times relative to saline and potentiated morphine-induced locomotion. Therefore, in M(5) knockout mice, blockade of either VTA muscarinic or nicotinic receptors increased locomotion, suggesting that in the absence of VTA M(5) receptors, VTA cholinergic inputs inhibit locomotion.

Carbachol injections into the intergeniculate leaflet induce nonphotic phase shifts.

Nonphotic phase shifts of the circadian clock in mammals are mediated by the intergeniculate leaflet (IGL) of the thalamus via a geniculohypothalamic projection to the suprachiasmatic nucleus. These shifts can be induced by arousing stimuli, such as wheel running, brain stimulation reward and foot shock. Because mesopontine cholinergic neurons are also activated by arousing stimuli, we tested the hypothesis that cholinergic input to the IGL mediates nonphotic phase shifts. Carbachol injected into the IGL of hamsters in their subjective day (CT8) induced phase advances similar to shifts that are induced by arousal at the same circadian time. Control injections of saline at CT8 did not advance phase similarly. Carbachol injections outside the IGL produced smaller shifts. Pre-injections of the muscarinic antagonist, atropine, reduced carbachol-induced phase advances relative to saline pre-injections. The results indicate that muscarinic input to the IGL can induce nonphotic phase shifts.

Reward and aversive stimuli produce similar nonphotic phase shifts.

Circadian rhythms in rodents respond to arousing, nonphotic stimuli that contribute to daily patterns of entrainment. To examine whether the motivational significance of a stimulus is important for eliciting nonphotic circadian phase shirts in Syrian hamsters (Mesocricetus auratus), the authors compared responses to a highly rewarding stimulus (lateral hypothalamic brain stimulation reward [BSR]) and a highly aversive stimulus (footshock). Animals were housed on a 14:10-hr light-dark cycle until test day, when they were given a 1-hr BSR session (trained animals) or a 1-mA electric footshock at 1 of 8 circadian times, and were maintained in constant dark thereafter. Both BSR pulses and footshock produced nonphotic phase response curves. These results support the hypothesis that arousal resulting from the motivational significance of a stimulus is a major factor in nonphotic phase shifts.

Role of cholinergic receptors in locomotion induced by scopolamine and oxotremorine-M.

Mesopontine cholinergic neurons activate dopamine neurons important for reward-seeking and locomotor activity. The present studies tested whether cholinergic receptor blockade in the ventral tegmental area (VTA) altered locomotion induced by scopolamine (3 mg/kg i.p.) or by oxotremorine-M (0.1 microg bilaterally in the VTA). It was predicted that cholinergic blockers in the VTA would attenuate these cholinergic-induced locomotor increases. Locomotor activity was increased by scopolamine and oxotremorine-M administration in all treatments. When dihydro-beta-erythroidine (DHBE), a nicotinic receptor antagonist, was applied in VTA prior to oxotremorine-M, locomotion was reduced to slightly above saline baseline levels, but atropine, a muscarinic antagonist, had no effect. This suggests that the locomotor effect of oxotremorine-M at this dose was mediated mainly via nicotinic, not muscarinic, receptors. Intra-VTA injections of DHBE, however, did not attenuate scopolamine-induced locomotion indicating that scopolamine-induced locomotion is not mediated mainly via VTA cholinergic receptors. In mutant mice with a deletion in the M5 muscarinic receptor gene, scopolamine-induced locomotion was increased versus wild type mice after scopolamine injection. This suggests that the M5 receptor has an inhibitory effect on scopolamine-induced locomotion.

Enhancement of electrically evoked startle-like responses by tetanic stimulation of the superior colliculus.

Single-pulse unilateral electrical stimulation of either the amygdala or the inferior colliculus elicited startle-like responses in chloral hydrate anesthetized rats. EMG responses to intracranial stimulation were recorded from the anterior biceps femoris muscles. The EMG responses were generally enhanced following unilateral tetanic stimulation of the deep layers of the superior colliculus, but the enhancement was stronger for amygdala sites than inferior colliculus sites. The enhancement of EMG responses to ipsilateral amygdala stimulation was much larger than that for contralateral amygdala stimulation and that for ipsilateral inferior colliculus stimulation. The enhancement of EMG responses to contralateral inferior colliculus stimulation was not significant. The present study provides a motor-output model for studying plasticity in the neural pathways mediating startle facilitation.

Tactile, acoustic and vestibular systems sum to elicit the startle reflex.

The startle reflex is elicited by intense tactile, acoustic or vestibular stimuli. Fast mechanoreceptors in each modality can respond to skin or head displacement. In each modality, stimulation of cranial nerves or primary sensory nuclei evokes startle-like responses. The most sensitive sites in rats are found in the ventral spinal trigeminal pathway, corresponding to inputs from the dorsal face. Cross-modal summation is stronger than intramodal temporal summation, suggesting that the convergence of acoustic, vestibular and tactile information is important for eliciting startle. This summation declines sharply if the cross-modal stimuli are not synchronous. Head impact stimuli activate trigeminal, acoustic and vestibular systems together, suggesting that the startle response protects the body from impact stimuli. In each primary sensory nucleus, large, second-order neurons project to pontine reticular formation giant neurons critical for the acoustic startle reflex. In vestibular nucleus sites, startle-like responses appear to be mediated mainly via the vestibulospinal tract, not the reticulospinal tract. Summation between vestibulospinal and reticulospinal pathways mediating startle is proposed to occur in the ventral spinal cord.

M5 muscarinic receptors are required for prolonged accumbal dopamine release after electrical stimulation of the pons in mice.

Midbrain dopamine neurons are activated directly by cholinergic agonists or by stimulation of the cholinergic neurons in the laterodorsal tegmental nucleus (LDT) of the pons in rats. In urethane-anesthetized mice, electrical stimulation of the LDT resulted in a rapid, stimulus-time-locked increase in dopamine release in the nucleus accumbens (NAc), followed several minutes later by a prolonged increase in dopamine release. In mutant mice with truncated M5 receptors, the prolonged phase of dopamine release was absent, but the initial, rapid phase of dopamine release was fully observed. We conclude that M5 muscarinic receptors on midbrain dopamine neurons mediate a prolonged facilitation of dopamine release in the NAc. These results imply that M5 muscarinic receptors play an important role in motivational behaviors driven by dopamine activity in the accumbens.